Foraster Cabrera, Arnau2015-04-132015-04-132015-01-21https://hdl.handle.net/2445/64965Treballs Finals de Grau d'Enginyeria Informàtica, Facultat de Matemàtiques, Universitat de Barcelona, Any: 2015, Director: Lluís Garrido OstermannThe back idea of this project consists of creating an automatized system which is able to forecast the result of the matches of the Spanish football league (LFP). It has been developed in different parts. One is related to request and process data from a website. Another part is used to train and test a classifier with the stored data using a set of supervised learning methods. This kind of classifiers are commonly called support vector machines (SVM). And the last one, create a web application to check and control part of the system. To achieve a high percentage of success, it’s very important to have enough data for each team. Also, it’s necessary to identify which features are relevant for the problem of classification. And adjusting the parameters of the classificator in a good way make a substantial difference.This document outlined in detail the process of design, analysis, development, implementation and results of this project.54 p.application/pdfcatmemòria: cc-by-sa (c) Arnau Foraster Cabrera, 2015codi: GPL (c) Arnau Foraster Cabrera, 2015http://creativecommons.org/licenses/by-sa/3.0/eshttp://www.gnu.org/licenses/gpl-3.0.ca.htmlAprenentatge automàticAlgorismesProgramariTreballs de fi de grauMineria de dadesFutbolMachine learningAlgorithmsComputer softwareBachelor's thesesData mininginfo:eu-repo/semantics/bachelorThesisinfo:eu-repo/semantics/openAccess